J Appl Electrochem (2010) 40:653–662
DOI 10.1007/s10800-009-0040-y
ORIGINAL PAPER
Electrochemical behavior of copper current collector
in imidazolium-based ionic liquid electrolytes
Chengxin Peng • Li Yang • Shaohua Fang •
Jixian Wang • Zhengxi Zhang • Kazuhiro Tachibana
Yong Yang • Shiyong Zhao
•
Received: 11 June 2009 / Accepted: 23 November 2009 / Published online: 10 December 2009
Ó Springer Science+Business Media B.V. 2009
Abstract The electrochemical behaviors of copper current
collector in 1-alkyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl] imide ionic liquid electrolytes were
investigated and compared with that in ethylene carbonate/
dimethyl carbonate solutions. Cyclic voltammetry results
showed that large oxidation–reduction current of the copper
foil appeared in ethylene carbonate/dimethyl carbonate
solutions, while a much smaller current in the room temperature ionic liquid electrolytes decreased gradually, indicating that the copper foil was anodically stable. Further
study by X-ray photoelectron spectroscopy analysis showed
that an unstable product was composed mainly of the carbonate and carbonyl species on the surface of the copper foil
after the electrochemical measurement in ethylene carbonate/dimethyl carbonate solutions, leading to the dissolution
of the copper foil. While a better passivating film from the
C. Peng
The Institute for Advanced Materials and Nano Biomedicine,
Tongji University, Shanghai 200092, China
L. Yang (&) S. Fang J. Wang Z. Zhang
School of Chemistry and Chemical Technology, Shanghai Jiao
Tong University, Shanghai 200240, China
e-mail: [email protected]
K. Tachibana
Department of Chemistry and Chemical Engineering,
Yamagata University, Yamagata 992-8510, Japan
Y. Yang
State Key Lab of Physical Chemistry of Solid Surface,
Xiamen University, Xiamen 361005, China
S. Zhao
Zhangjiagang Green Battery Material Research Institute Co. Ltd,
Zhangjiagang 215600, China
reduction of the anions in the room temperature ionic liquid
electrolytes covered the surface of copper foil and protected
the copper foil from being oxidized even in a higher potential. These results indicate that the use of room temperature
ionic liquid electrolytes can improve the stability of copper
current collector in the advanced lithium ion batteries.
Keywords Lithium ion battery Copper current
collector Corrosion Ionic liquid Electrolytes
1 Introduction
Copper (Cu) is commonly used as the anodic current collector for negative electrodes in lithium ion batteries. It
plays an important role in exchanging electrons between
the electrode materials and the external circuit. However,
Cu current collector may be subjected to the corrosion
during the charge/discharge processes, which can degrade
the performance and worsen the safety of lithium ion batteries. P. Arora et al. [1] have summarized the capacity
fading mechanisms in lithium ion batteries including the
dissolution of Cu current collector. Further research [2] by
Yang et al. showed that the dissolved Cu ions from the
current collector precipitated as Cu oxides on the carbon
surface, which blocked the normal intercalation of lithium
ions and led to the capacity fading of lithium ion batteries.
Besides, Aurbach and Cohen [3] suggested that copper
in LiAsF6-PC was not completely inert at open-circuit
potential (OCP) and contaminants might oxidize the copper. Also, Zhao et al. [4] have reported that the dissolution
of copper in ternary organic mixture containing LiPF6
appeared at about 3.5 V versus Li?/Li, which was less than
0.5 V above the OCP of the electrode. Although other
study has described that the corrosion of copper could not
123
2 Experimental
LiTFSI salt (99.9%, Morita) was dried under vacuum at
120 °C for 24 h prior to use. Room temperature ionic
123
H3C
N
+
be realized on the negative electrode on account of that the
upper limit of its potential is estimated to be 1.5 V versus
Li?/Li in most practical uses of the commercial lithium ion
batteries [5], the dissolution of Cu current collector is still
inevitable to occur during the overdischarged process [4],
especially for lithium ion batteries with high voltage
cathodic materials [6], the copper current collector for
negative electrodes may reach a potential high enough to
be corroded at the end of the discharge process. Therefore,
it is necessary to investigate the stability of copper current
collector in the electrolytes which is used to improve the
performance of lithium ion battery.
Organic-based electrolytes have been widely used in
lithium ion batteries. Owing to their flammability and volatility, their use in lithium ion battery is limited, especially for
those advanced lithium ion batteries used as the power
source of the electric vehicle (EV) and other large-scale
power systems [7, 8]. Recently, room temperature ionic
liquids (RTILs) have received considerable attentions due to
their excellent properties, such as high thermal stability, nonflammability, low vapor pressure, and wide electrochemical
window [9–14], etc. Among these RTILs, the imidazoliumbased room temperature ionic liquids are most commonly
used owing to their relatively easy synthesis, considerable
ion conductivity as 10-2 S cm-1 and so on. Besides, lithium
bis[(trifluoromethyl)sulfonyl] imide [LiN(CF3SO2)2, LiTFSI] salt has great thermal, hydrolytic stability, good conductivity, and high electrochemical stability. These RTILs
containing LiTFSI can be potentially used as a good candidate for lithium ion batteries [15–17]. However, up to now,
there were few reports studying the current collector
behaviors in these RTIL electrolytes. In the previous report,
we have showed that the Al cathode current collector was
stable in a series of 1-alkyl-3-methylimidazolium bis[(trifluoromethyl) sulfonyl] imide ionic liquid electrolytes [18],
whereas the electrochemical behaviors of the Cu anode
current collector in these RTIL electrolytes are still unclear.
In this article, the electrochemical behaviors of the Cu
current collector were studied in 1-alkyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl] imide ionic liquid
electrolytes. The results were compared with that in ethylene carbonate (EC)/dimethyl carbonate (DMC) solutions
to find more suitable electrolytes with better compatibility
between the RTILs and Cu current collector. In this study,
cyclic voltammetry (CV), scanning electron microscopy
(SEM), and X-ray photoelectron spectroscopy (XPS) were
used for characterization.
J Appl Electrochem (2010) 40:653–662
S
O
CH2CH3
H3C
N
-
N
S
N
CH2CH2CH2CH3
O
O
O
O
F3C
N
+
654
CF3
F3C
O
[EtMeIm]N(SO2CF3)2
S
-
N
O
S
CF3
O
[BuMeIm]N(SO2CF3)2
EMI-TFSI
BMI-TFSI
Fig. 1 Chemical structures of the RTILs
liquids based on TFSI- anion, 1-ethyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl] imide (EMI-TFSI), and
1-butyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl] imide (BMI-TFSI) were prepared in our laboratory
according to the procedures in the previous studies [19,
20]. Before making solutions, these RTILs were treated
with molecular sieves (4 Å) for 2 days to dehydrate so that
water content in them was less than 30 ppm which was
determined by Karl Fisher titration method (Fig. 1 shows
their chemical structures). LiTFSI salt was dissolved in
EMI-TFSI, BMI-TFSI, and EC/DMC (1:1 in vol., 98%,
Zhangjiagang Guotai Huarong New Chemical Materials
Co. Ltd.) to make 1 mol dm-3 (M) solutions, respectively.
The electrochemical behaviors of the Cu current collector were investigated using CHI 604B electrochemical
work-station (CH Instruments, USA). A standard cell
(inner volume: 10 cm3) with three isolated electrodes was
applied in this experiment. A specimen of Cu foil (13 lm
thickness, 99.9%) was cut into small pieces as tape electrodes with dimensions of 50 mm 9 50 mm exposed to the
electrolytes. After Cu foil was degreased in acetone for
30 min and dried at room temperature under vacuum, it
was electropolished at an anodic constant potential of
2.0 V in a phosphoric acid solution for about 15 min. It
was then rinsed thoroughly in a 10% phosphoric acid
solution and deionized water. Finally, Cu foil was dried at
room temperature under vacuum for 48 h. The treated
copper foil, lithium foil, and lithium foil were used as
working electrode, reference, and counter electrodes,
respectively. Cyclic voltammetry of the copper foil in each
electrolytes mentioned above was performed for three
times with a potential scan rate of 10.0 mV s-1. Before the
experimental, the electrochemical windows of the electrolytes were tested using linear sweep voltammetry (LSV)
under a scan rate of 10.0 mV s-1. In this scan, a platinum
(Pt) disk with 2 mm diameter and two pieces of Li foils
were used as working electrode, reference, and counter
electrodes, respectively. In this article, all potentials were
relative to the standard potential of Li?/Li. The electrolytes preparation processes and the electrochemical
measurements were conducted in UNiLab glove box
J Appl Electrochem (2010) 40:653–662
655
([O2] \ 20 ppm, [H2O] \ 20 ppm) filled with argon (Ar)
atmosphere at ambient temperature.
After the electrochemical measurements, the copper
foils were well rinsed by the dimethyl carbonate (DMC)
solvent and then dried at room temperature under vacuum
for 24 h. These copper foils were then transferred in a
portable sealed carrier from the Ar-filled glove box to the
sample holder of SEM (LEO 1530, Oxford Instrument)
chamber to observe their surface morphology.
X-ray photoelectron spectroscopy (XPS) measurement
was applied to analyze the composition on the surface of
copper foils after the electrochemical measurements. A
sample preparation was conducted using a similar method
with the SEM observation. The analysis chamber of XPS
(Quantum 2000, Physical Electronics) was kept at
5 9 10-8 Pa and Al Ka line was used as an X-ray source
which operated at 8 mA and 2 kV to obtain the spectra of
each element existing on the surface of copper foils after the
electrochemical measurements.
3 Results and discussion
3.1 Electrochemical behaviors of Cu foil electrode
in the RTIL electrolytes and the EC/DMC solutions
In order to study the electrochemical behaviors of Cu foil
in the electrolytes, the electrochemical windows of EC/
DMC containing 1 M LiTFSI and the RTILs were tested at
the beginning. As shown in Fig. 2, the reduction and the
oxidation potentials of the RTILs occurred at around 1.5
and 6.0 V, while the corresponding potentials of the EC/
DMC solutions were found at about 0.5 and 5.7 V,
respectively. These results were in good agreement with
Current density /mA cm-2
1mA cm-2
1M LiTFSI / EC+DMC
EMI-TFSI
BMI-TFSI
0
1
2
3
4
5
6
7
+
Potential / V vs. Li / Li
Fig. 2 Linear sweep voltammetry on platinum at EC/DMC containing 1 M LiTFSI and RTILs (sweep rate: 10.0 mV s-1)
the previous reports [18, 21]. Almost no current was
observed obviously in the potential range of 1.5–5.5 V for
above electrolytes, which indicated that no more electrolytes degradation considerably occurred.
Cyclic voltammetry (CV) was performed in the potential
range of 1.5–5.5 V to investigate the electrochemical
behaviors of the Cu current collector in the electrolytes.
Figure 3 showed the anodic behaviors of the Cu foil in EC/
DMC solutions and the RTILs containing 1 M LiTFSI.
During the anodic potential sweep from 3.0 to 3.8 V, no
more electrolytes degradation occurred obviously. While
for the EC/DMC solutions shown in Fig. 3a, a sharp continuous current rise at about 3.5 V, which was ascribed to
the oxidation of the Cu foil. In the reverse scan, the
reduction of the Cu ion was observed at around 3.4 V [4,
22]. Similar results also appeared in LiPF6 and LiClO4based electrolytes [4, 5]. These phenomena showed that
copper current collector was prone to oxidation in the
organic electrolytes at the potentials of around 3.5 V.
Therefore, the potentials of negative electrodes in the
batteries should be limited to about 3.5 V. Besides, the
current density of the Cu foil in the EC/DMC solutions
slightly increased during the following sweeps, which
indicated that the Cu foil was not stable and its corrosion
occurred in the EC/DMC solutions. It is disadvantageous
because the dissolved copper will plate on surface of the
carbon anode during the charge/discharge processes,
leading to degradation of the performance of lithium ion
batteries. For the RTIL electrolytes shown in Fig. 3b and c,
a weaker anodic current at about 3.4 V was observed,
compared with that of the EC/DMC solutions. It rapidly
decreased during the following cycles, which suggested
that the Cu foil was stable in the RTIL electrolytes. The
reason was that a protective film might form on the surface
of Cu foil in the RTIL electrolytes during the potential
sweeps and it suppressed the oxidation of the Cu foil as a
result. The stability of the Cu current collector in the RTIL
electrolytes can improve the safety of lithium ion batteries
during the charge/discharged processes, especially at an
over-discharged state. For example, when a practical lithium ion battery containing LiCoO2 was over-discharged to
*0.4 V, copper corrosion would occur in EC/DMC solutions while not in the RTIL electrolytes since the Eanode
reached *3.4 V, based on the fact that OCP of fully discharged LixCoO2 is *3.8 V [23].
Cyclic voltammetry was also performed to investigate
the reduction of the electrolytes on the surface of the Cu
foil during the cathodic scan. The voltammetric response of
the Cu foil in EC/DMC solutions from 3.0 to 0.0 V was
shown in Fig. 4a, obvious reduction peaks at around 0.3
and 0.8 V appeared, which were attributed to the reduction
of the solvents in the electrolytes [24]. While the other
reduction peaks at about 1.2 V was assigned to the
123
656
J Appl Electrochem (2010) 40:653–662
(a)
(a)
0.5
1st cycle
2nd cycle
3rd cycle
2
Current density / mA cm-2
Current density / mA cm-2
3
1
0
3.0
3.2
3.4
3.6
1st cycle
2nd cycle
3rd cycle
0.0
3.8
Potential / V vs. Li+/Li
0.0
0.5
1.0
1.5
2.0
2.5
3.0
2.5
3.0
2.5
3.0
Potential / V vs. Li+/Li
(b)
1st cycle
2nd cycle
3rd cycle
0.5
Current density / mA cm-2
Current density / mA cm-2
(b)
0.02
0.01
0.00
3.0
3.2
3.4
3.6
1st cycle
2nd cycle
3rd cycle
0.0
3.8
Potential / V vs. Li+/Li
0.0
0.5
1.0
1.5
2.0
Potential / V vs. Li+/Li
(c)
0.02
1st cycle
2nd cycle
3rd cycle
0.5
Current density / mA cm-2
Current density / mA cm-2
(c)
0.01
0.00
3.0
3.2
3.4
3.6
1st cycle
2nd cycle
3rd cycle
0.0
3.8
Potential / V vs. Li+/Li
0.0
0.5
1.0
1.5
2.0
+
Fig. 3 Cyclic voltammograms for Cu foil electrode in 1 M LiTFSI–
EC/DMC (a); 1 M LiTFSI–EMI-TFSI (b); 1 M LiTFSI–BMI-TFSI
(c) (sweep potential range: 3.0–3.8 V; scan rate: 10.0 mV s-1)
reduction of residual moisture in the electrolytes [25].
When the potential reached 0.0 V, a reduction peak was
observed, which was followed by a very small oxidation
peak at about 1.0 V and a broad peak near 1.5 V in the
reverse scan. Compared with that in the RTIL electrolytes
123
Potential / V vs. Li /Li
Fig. 4 Cyclic voltammograms for Cu foil electrode in 1 M LiTFSI–
EC/DMC (a); 1 M LiTFSI–EMI-TFSI (b); 1 M LiTFSI–BMI-TFSI
(c) (sweep potential range: 3.0–0.0 V; scan rate: 10.0 mV s-1)
shown in Fig. 4b and c, the oxidation peak at around 1.0 V
was absent in the RTIL electrolytes, indicating that this
oxidation peak might be ascribed to the reverse reaction of
the reduction products of the solvents at 0.3 and 0.8 V.
J Appl Electrochem (2010) 40:653–662
Current density / mA cm-2
(a)
1st cycle
2nd cycle
3rd cycle
0.5
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
3.5
4.0
3.5
4.0
+
Potential / V vs. Li /Li
Current density / mA cm-2
(b)
0.5
1st cycle
2nd cycle
3rd cycle
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
+
Potential / V vs. Li /Li
(c)
Current density / mA cm-2
Besides, the other typical peaks at 0.0 V and *1.5 V
appeared during the potential sweeps in the RTIL electrolytes. The peak at 0.0 V might not be assigned to the
reduction of the lithium ion because the dissolution of
lithium in the EMI-TFSI electrolytes happened at -0.5 V.
Moreover, the oxidation peak of lithium ion at the potential
of *0.25 V did not present in the reverse scan [26].
Consequently, these two peaks might be related to the
reduction and the reverse reaction of the reduction products
of N(SO2CF3)2 anion which existed in both electrolytes.
The peak at around 2.0 V in both kinds of electrolytes
systems was ascribed to the oxidation of the Cu to Cu?
[27]. The different CV features of the Cu foil in the RTIL
electrolytes and EC/DMC solutions suggested that almost
no electrolytes degraded greatly during the potential
sweeps in the RTIL electrolytes, which may be useful to
enhance the safety of lithium ion batteries, especially at an
over-charged state.
In oreder to further evaluate the influence of the
reduction of the electrolytes on the electrochemical
behaviors of the Cu current collector, the CVs of the Cu
foil both in the EC/DMC solutions and in the RTIL electrolytes were carried out during the potential swept from
0.0 to 3.8 V. In the EC/DMC solutions, it was observed
that the anodic peak of the Cu foil at the potential around
3.5 V decreased gradually with the continual cycles
(Fig. 5a), which was opposite to the continuous increasing
CV features of the Cu foil in the same electrolytes during
the potential swept from 3.0 to 3.8 V. According to the
results from the preliminary experiments, no decreasing
tendency of the Cu oxidation peak near 3.5 V took place
during the cathodic scan with different switch potentials
until the reduction peaks appeared at around 0.3 and 0.8 V.
These phenomena revealed that some amounts of electrolytes reduced and formed some protection on the Cu foil.
These passivating film attenuated the further oxidation of
the Cu foil and thus the oxidation current of the Cu foil at
around 3.5 V decreased. Zhao et al. [4] have also showed
similar results that the passivating film formed on the
copper foil and gave some protection to suppress its oxidation in ternary organic mixture containing LiPF6. Figure 5b and c showed the CVs of the Cu foil in the RTIL
electrolytes. A weaker oxidation current of the Cu foil
electrode in both RTIL electrolytes was observed compared with that in the EC/DMC solutions, indicating that a
better film formed on the Cu foil electrode which inhibited
the further oxidation of the Cu foil. These results confirmed
that the reduction of the electrolytes during the potential
sweeps might provide some protection for Cu current
collector for lithium ion batteries.
In order to examine the stability of the protection
formed on the Cu foil during the potential sweeps in
EC/DMC solutions and the RTIL electrolytes, the potential
657
0.5
1st cycle
2nd cycle
3rd cycle
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
+
Potential / V vs. Li /Li
Fig. 5 Cyclic voltammograms for Cu foil electrode in 1 M LiTFSI–
EC/DMC (a); 1 M LiTFSI–EMI-TFSI (b); 1 M LiTFSI–BMI-TFSI
(c) (sweep potential range: 0.0–3.8 V; scan rate: 10.0 mV s-1)
swept positively to 5.0 V. In the EC/DMC solutions, an
oxidation current of the Cu foil increased rapidly at about
4.0 V in the first cycle (Fig. 6a), implying the dissolution
of copper occurred on the Cu foil. However, a higher
123
658
J Appl Electrochem (2010) 40:653–662
Current density / mA cm-2
15
(a)
1st cycle
2nd cycle
3rd cycle
10
5
0
0
1
2
3
4
5
Potential / V vs. Li+/Li
Current density / mA cm-2
(b)
0.5
1st cycle
2nd cycle
3rd cycle
0.0
0
1
2
3
4
5
Potential / V vs. Li+/Li
(c)
Current density / mA cm-2
0.5
1st cycle
2nd cycle
3rd cycle
3.2 Morphology of Cu foil electrode after
the electrochemical measurement
0.0
0
1
2
3
4
5
+
Potential / V vs. Li /Li
Fig. 6 Cyclic voltammograms for Cu foil electrode in 1 M LiTFSI–
EC/DMC (a); 1 M LiTFSI–EMI-TFSI (b); 1 M LiTFSI–BMI-TFSI
(c) (sweep potential range: 0.0–5.0 V; scan rate: 10.0 mV s-1)
oxidation current occurred during the reverse potential
scans showed that the process was accompanied by a
corrosion step on the surface of the Cu foil. In the continuous cycles, the magnitude of the oxidation current
123
increased and the corrosion potential simultaneously shifted negatively to a lower potential. These behaviors suggested that a less effective protection film formed on the
surface of the Cu foil. Thus, it was prone to the breakdown
and the dissolution of the Cu foil appeared. In the experiment, it was observed that the Cu foil was broken down and
large amounts of pits appeared on its surface after five
potential scans, indicating that serious Cu foil corrosion
occurred during the potential sweeps in the EC/DMC
solutions. Although the reduction of the electrolytes had
provided the limited protection for the Cu foil, the protective film was not stable and not enough to protect the Cu
foil from being eroded at the higher voltage. Whereas, in
the RTIL electrolytes, it was found that a weaker oxidation
current of the Cu foil presented and it decreased with the
following cycles. The oxidation current was less than
0.05 mA cm-2, much smaller than that in the EC/DMC
solutions. This is because the stable protective film on the
surface of the Cu foil, which came from the reduction of
the RTIL electrolytes during the cathodic scan, inhibited
the dissolution of the Cu foil even the potential up to 5.0 V.
Besides, similar shape of cyclic voltammograms of the Cu
foil suggested that the electrochemical behaviors on the
Cu foil were essentially same in both RTIL electrolytes.
Therefore, it can be concluded that the Cu foil is completely passivated in the RTIL electrolytes which can resist
higher anodic potential. The high stability of copper current collector in the RTIL electrolytes can be available to
improve the safety of lithium ion batteries during the
overcharged process, especially for the stacks of batteries
during the reverse-charged state when some individual cell
encountered the mistaken way in the assemble processes.
In order to investigate the formation of the passivating film
more clearly, the morphology of the Cu foil was observed
by SEM after multiple CV scans in the EC/DMC solutions
and the RTIL electrolytes. In addition, the morphology of a
raw copper foil (Fig. 7d) was also taken as a reference.
Figure 7a–c showed the SEM images of the Cu foil after
the potential sweeps from 0.0 to 3.8 V in the EC/DMC
solutions and the RTIL electrolytes. As shown in Fig. 7a,
some pits of 5–15 lm size were observed on the surface of
the Cu foil, indicating that the dissolution of the Cu foil
took place in the EC/DMC solutions, which led to the rise
of anodic current during the potential swept to about 3.5 V.
On the other hand, some solid products still formed on the
Cu foil surface after the potential scans in the EC/DMC
solutions. These compounds came from the reduction of
the electrolytes during the potential sweeps, and played a
role in attenuating the further corrosion of the Cu foil and
J Appl Electrochem (2010) 40:653–662
659
Fig. 7 SEM images of Cu foil
electrode surface after the
potential scanning from 0.0 to
3.8 V in 1 M LiTFSI–EC/DMC
(a); 1 M LiTFSI–EMI-TFSI (b);
1 M LiTFSI–BMI-TFSI (c) and
raw copper foil (d)
induced the decrease of the oxidation current at about
3.5 V during the potential sweeps from 0.0 to 3.8 V.
However, this limited protective film was not stable enough
to protect the Cu foil from its corrosion over higher
potential. While for both EMI-TFSI and BMI-TFSI electrolytes, morphologies of the copper foils (Fig. 7b and c)
showed that a deposited products film was formed on the
Cu foil substrate after the potential sweeps, which protected the Cu foil electrode from its further oxidation. As a
result, almost no obvious oxidation current of the Cu foil
was observed from the above CV response. This obeservation indicated that the Cu foil surface was wholly passivated in both RTIL electrolytes after the electrochemical
measurements. Based on the above results, it can be concluded that the above SEM images are in good agreement
with the CV results. Although the protection behaviors for
the Cu foil electrode were all observed in EC/DMC solution and RTIL electrolytes, higher oxidation current of the
Cu foil in the EC/DMC electrolyte than that in the RTIL
electrolytes, indicating that the dissolution of the Cu foil
still existed in EC/DMC solution and it competed with the
limited protection of the passivating film on the Cu foil.
Once the protection for the Cu foil did not suppress the
oxidation of the Cu foil, the copper corrosion happened. In
comparison, a better passivating film formed in both EMITFSI and BMI-TFSI electrolytes, which was available to
protect the Cu foil from its further oxidation.
3.3 Character of the Cu foil surface after
the electrochemical measurement
In order to understand the complicated mechanism of
corrosion and passivation processes, the passivating film on
the Cu foil was analyzed by XPS measurement after the
electrochemical measurement in EC/DMC solutions and
the RTIL electrolytes. Additionally, high-resolution region
spectra were obtained by a least-square fitting of model
curve (XPSPEAK Version 4.0 software) after the removal
of Shirley-type background.
Figure 8 showed the XPS spectra of C1s, O1s, N1s, F1s,
Cu2p, and S2p for the Cu foil after the potential sweeps from
0.0 to 3.8 V in three electrolytes, and the XPS binding
energies were summarized in Table 1. According to the XPS
spectra, the most difference between the EC/DMC solutions
and the RTIL electrolytes was the signal of C1s and O1s
spectra. As for the C1s spectra, the presence of species
associated with C–C/C–H at 285.0 eV, C–O at 286.6 eV,
and CO32- or ROCO2Li at 289.0 eV [28] were found in all
three electrolytes. However, the intensity of C1s peak at
289.0 eV was higher in the EC/DMC solutions than that in
the RTIL electrolytes, showing that almost no oxalates presented on the Cu foil surface after the potential sweeps in
both RTIL electrolytes. Several mechanisms in following
Eqs. 1–3 have been reported to explain their formation [29].
In the EC/DMC solutions, some solvents (such as EC and
123
660
J Appl Electrochem (2010) 40:653–662
Fig. 8 XPS spectra for Cu foil electrode after the potential scanning from 0.0 to 3.8 V in 1 M LiTFSI–EC/DMC (a); 1 M LiTFSI–EMI-TFSI
(b); 1 M LiTFSI–BMI-TFSI (c)
DMC) and their reduction products (Li2CO3 and CH3OCO2Li) accounted for the appearance of three C1s peaks,
while only the C1s peaks at 285.0 and 286.6 eV suggested the
residual of DMC solvent after the rising process on the surface of the Cu foil in the RTIL electrolytes.
EC: ðCH2 OÞ2 CO þ 2e þ 2Liþ ! Li2 CO3 þ C2 H4
ð1Þ
DMC: CH3 OCO2 CH3 þ 2e þ 2Liþ ! Li2 CO3 þ C2 H6
ð2Þ
DMC: 2CH3 OCO2 CH3 þ 2e þ 2Liþ
! 2CH3 OCO2 Li þ C2 H6
ð3Þ
Peaks presented in the O1s high-resolution region spectrum were identified as CO32- or SO2 at 532.5 eV [30], C=O
at 531.5 eV and Cu2O or CuO at 530.2 eV [31]. These peaks
were all observed on the surface of the Cu foil after the
electrochemical measurement in three electrolytes. However, a newly XPS spectra peak at 528.5 eV, which was
ascribed to Li2O [32], appeared on the Cu foil surface during
the potential sweeps in both EMI-TFSI and BMI-TFSI
electrolytes. These results indicated that a significant quantity of Li2O formed on the surface of the Cu foil after the
electrochemical measurement in the RTIL electrolytes.
Although the presence of Li2O may increase the internal
resistance which will adversely affect the battery performance, the stable Li2O formed on the Cu foil surface may
improve the stability of the protective film, which will
facilitate Cu current collector to resist higher voltage.
In addition to the C1s and O1s spectra, other spectra also
showed the difference among the three electrolytes. The F1s
high-resolution region spectra presented two peaks at 685.0
and 687.8 eV, which were identified to LiF and –CF3,
123
respectively [28]. The N1s high-resolution region spectrum
displayed two peaks corresponding to nitride at 397.5 eV
and NSO2CF3- at 399.7 eV [33], and the S2p high-resolution
region spectrum showed the presence of a number of peaks
which were mostly likely associated with elemental sulfur
and lithium sulfide from 160–165 eV, oxidized sulfur
species at 167.0 and 168.2 eV, and SO2CF3- species at
169.0 eV [30]. By comparing the XPS spectra, it can be
observed that higher intensity of F1s spectra peak at 687.8 eV
and N1s spectra peak at 399.7 eV presented in the RTIL
electrolytes, showing that a larger amount of species containing –CF3 and NSO2CF3- on the Cu foil. In addition, the
content of the S2p spectra at 169.0 eV assigned to SO2CF3species increased in the RTIL electrolytes. The formation of
the passivating products on the Cu foil was explained by
several mechanisms [30] in the following Eqs. 4–7. Based on
the above results, it was concluded that more amounts of
Li2O, CF3SO2Li, and CF3SO2NLi species were presented on
the surface of the Cu foil after the potential sweeps in the
RTIL electrolytes, the majority of which appeared to be the
reduction products of the N(SO2CF3)2 anion.
þ
NðSO2 CF3 Þ
2 þ ne þ ðn þ 1ÞLi
! Li3 N þ Li2 S2 O4 þ LiF þ C2 Fx Liy
NðSO2 CF3 Þ
2
ð4Þ
þ 2e þ 3Liþ ! Li2 NSO2 CF3 þ CF3 SO2 Li
ð5Þ
Li2 S2 O4 þ 6e þ 6Liþ ! 2Li2 S þ 4Li2 O
ð6Þ
Li2 S2 O4 þ 4e þ 4Liþ ! Li2 SO3 þ Li2 S þ Li2 O
ð7Þ
The XPS Cu2p spectra were also investigated to confirm
the passivating film on the Cu foil substrate. In the Cu2p 3/2
spectra, two obvious peaks for cuprous ion (Cu?) at about
–
932.4
168.0
161.2
400.0
397.4
687.8
684.5
532.5
531.2
530.2
527.7
–
284.4
(c) 1 M LiTFSI/BMI-TFSI
286.2
933.2
932.9
931.3
931.8
168.3
167.2
160.8
162.4
399.0
399.1
684.2
532.5
532.5
531.2
531.2
530.2
530.2
–
528.0
–
288.2
284.2
(b) 1 M LiTFSI/EMI-TFSI
286.2
284.3
(a) 1 M LiTFSI/EC ? DMC
286.5
684.3
687.2
397.5
661
397.5
3/2
Cu2p
S2p
N1s
F1s
O1s
C1s
Binding energy (eV)
Table 1 Summary of XPS binding energies (eV) for Cu foil electrode after the potential scanning from 0.0 to 3.8 V in 1 M LiTFSI–EC/DMC (a); 1 M LiTFSI–EMI-TFSI (b); 1 M LiTFSI–
BMI-TFSI (c)
J Appl Electrochem (2010) 40:653–662
932.0 eV and cupric ion (Cu2?) at around 933. 4 eV were
obtained on the surface of the Cu foil after the anodic scan.
However, the signal of the Cu2? spectra diminished for the
EC/DMC solutions and the EMI-TFSI electrolytes, and it
was absent in the BMI-TFSI electrolytes. As mentioned in
the previous reports [34], Cu? is more likely formed than
Cu2? in a non-aqueous environment. Other research [5] also
confirmed that the oxidized copper existed as a cuprous ion
(Cu?) in the organic electrolytes. Consequently, the
protective film on the Cu foil surface mainly consisted of
cuprous ion (Cu?) species such as Cu2O, Cu2CO3, or
Cu2(COO)2 after the anodic sweeps in the three electrolytes,
whereas the presence of cupric ion (Cu2?) in the protective
film may relate to the residual moisture in the electrolytes.
Overall, the XPS analysis results showed that the
products on the surface of the Cu foil after the electrochemical measurement in three electrolytes was mainly
composed of lithium salts and the cupreous species. In
addition, the results indicated a clearly different electrochemical behaviors of the Cu foil between the LiTFSI–EC/
DMC solutions and the LiTFSI–RTIL electrolytes. Carbonate and carbonyl species were present on the Cu foil
electrode surface after the electrochemical measurement in
the EC/DMC solutions and absent in the RTIL electrolytes.
While a larger amount reduction products of the
N(SO2CF3)2 anion formed on the surface of the Cu foil in
the RTIL electrolytes than that in the EC/DMC solutions.
These results were explained as that the solvents (EC and
DMC) were reduced earlier than the N(SO2CF3)2 anion,
which contributed to appearance of the main protective
film on the surface of the Cu foil in the EC/DMC solutions,
while the reduction products of the N(SO2CF3)2 anion
composed the passivating film on the surface of the Cu foil
after the electrochemical measurement in the RTIL electrolytes. With the above CV and XPS results of the Cu foil
after the potential swept to 5.0 V in two kinds of electrolytes, it can be concluded that these carbonate and carbonyl
species on the surface of the Cu foil in the LiTFSI–EC/
DMC solutions was not stable over 5.0 V and thus the
dissolution of the Cu foil appeared, while the reduction
products of the N(SO2CF3)2 anion formed on the surface
of the Cu foil after the potential sweeps in the RTIL
electrolytes, providing a effective protection for Cu current
collector to resist higher voltage for advanced lithium ion
batteries.
4 Conclusion
The electrochemical behaviors of the copper foil in
1-alkyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl] imide room temperature ionic liquids (RTILs) and
EC/DMC solutions containing LiTFSI were investigated. It
123
662
was found that the Cu foil displayed the oxidation–reduction behaviors at about 3.5 V in the EC/DMC solutions
while it was passivated in the RTIL electrolytes. Moreover,
the reduction of the electrolytes was also studied, showing
that the reduction of the electrolytes provided a limited
protection for the Cu foil after the electrochemical measurement in EC/DMC solutions, and it was not stable
enough to suppress the further oxidation of the Cu foil
when the potential swept to high potentials. In comparison,
a better passivating film covered on the surface of the Cu
foil in the RTIL electrolytes, which protected the Cu foil
from being oxidized. Further research by XPS analysis
showed that the solid products was mainly composed of the
reduction products of the electrolytes such as carbonate and
carbonyl species on the surface of the Cu foil after the
electrochemical measurement in the EC/DMC solutions.
Those solutions were not stable at the potential up to 5.0 V
and thus the dissolution of the Cu foil appeared. Whereas a
better passivating film, coming from the reduction of
(CF3SO2)2N- anion in the RTIL electrolytes, coated on the
Cu foil surface wholly to remarkably inhibit its further
oxidation and provided an effective protection for the
copper foil to resist higher voltage. These results indicated
that a good film-coating behavior of the Cu foil presented
in the RTIL electrolytes, which can improve the stability of
copper current collector for advanced lithium ion batteries.
Acknowledgments The authors thank the research center of analysis and measurement of Shanghai Jiao Tong University for the help
on the SEM measurements. This study was sponsored by the National
Grand Fundamental Research 973 Program of China under Grant No.
2006CB202600, the National High Technology Research and
Development Plan of China under Grant No. 2007AA03Z222.
References
1.
2.
3.
4.
Arora P, White RE, Doyle M (1998) J Electrochem Soc 145:3647
Yang L, Takahashi M, Wang B (2006) Electrochim Acta 5:3228
Aurbach D, Cohen Y (1997) J Electrochem Soc 144:3355
Zhao M, Kariuki S, Dewald HD, Lemke FR, Staniewicz RJ,
Plichta EJ, Marsh RA (2000) J Electrochem Soc 147:2874
5. Kawakita J, Kobayashi K (2001) J Power Sources 101:47
6. Hyams TC, Go J, Devine TM (2007) J Electrochem Soc
154:C390
7. Liaw BY, Dubarry M (2007) J Power Sources 174:76
123
J Appl Electrochem (2010) 40:653–662
8. Terada N, Yanagi T, Arai S, Yoshikawa M, Ohta K, Nakajima N,
Yanai A, Arai N (2001) J Power Sources 100:80
9. Blanchard LA, Hancu D, Beckman EJ, Brennecke JF (1999)
Nature 399:28
10. MacFarlane DR, Huang JH, Forsyth M (1999) Nature 402:792
11. Endres F, Abedin SZE (2006) Phys Chem Chem Phys 82:101
12. Zhou ZB, Matsumoto H, Tatsumi K (2004) Chem Eur J 10:6581
13. Hirao M, Sugimoto H, Ohno H (2000) J Electrochem Soc
147:4168
14. McEwen AB, Ngo HL, LeCompte K, Goldman JL (1999)
J Electrochem Soc 146:1687
15. Garcia B, Lavallée S, Perron G, Michot C, Armand M (2004)
Electrochim Acta 494:583
16. Seki S, Kobayashi Y, Miyashiro H, Ohno Y, Usami A, Mita Y,
Kihira N, Watanabe M, Terada N (2006) J Phys Chem B
110:10228
17. Seki S, Ohno Y, Kobayashi Y, Miyashiro H, Usami A, Mita Y,
Tokuda H, Watanabe M, Hayamizu K, Tsuzuki S, Hattori M,
Terada N (2007) J Electrochem Soc 154:A173
18. Peng C, Yang L, Zhang Z, Tachibana K, Yang Y (2007) J Power
Sources 173:510
19. Zhang Z, Gao X, Yang L (2005) Chin Sci Bull 50:2005
20. Bonhôte P, Dias A, Papageorgiou N, Kalyanasundaram K,
Grältzel M (1996) Inorg Chem 35:1168
21. Garcia B, Armand M (2004) J Power Sources 132:206
22. Cui Q, Dewald HD (2005) Electrochim Acta 50:2423
23. Staniewicz RJ, Romero A, Broussely M, Perton F, Labat J (1992)
In: Proceedings of the 35th International Power Sources Symposium. Institute of Electrical and Electronics Engineers, Inc.,
Piscataway, NJ, p 332
24. Zhang X, Kostecki R, Richardson TJ, Pugh JK, Ross PN Jr (2001)
J Electrochem Soc 148:A1341
25. Aurbach D, Talysosef Y, Markovsky B, Markevich E, Zinigrad E,
Asraf L, Gnanaraj JS, Kim H (2004) Electrochim Acta 50:247
26. Zhao L, Yamaki J -ichi, Egashira M (2007) J Power Sources
174:352
27. Choe HS, Carroll BG, Pasquariello DM, Abraham KM (1997)
Chem Mater 93:69
28. Leroy S, Martinez H, Dedryvère R, Lemordant D, Gonbeau D
(2007) Appl Surf Sci 253:4895
29. Dedryvère R, Laruelle S, Grugeon S, Gireaud L, Tarascon J-M,
Gonbeau D (2005) J Electrochem Soc 152:A689
30. Howlett PC, Brack N, Hollenkamp AF, Forsyth M, MacFarlane
DR (2006) J Electrochem Soc 153:A595
31. Moulder JF, Stickle WF, Sool PE, Bomber KD (1992) Handbook
of X-ray Photoelectron Spectroscopy, Perkin-Elmer, Eden Prairie
32. Kanamura K, Tamura H, Shiraishi S, Takehara Z-I (1995)
J Electroanal Chem 394:49
33. Dedryvère R, Leroy S, Martinez H, Blanchard F, Lemordant D,
Gonbeau D (2006) J Phys Chem B 110:12986
34. IUPAC Chemical Data Series-No. 22 (1979) Stability constants
of Metal-Ion Complexes, Part B, Organic Ligands, Compiled by
Douglas D. Perrin, Pergamon Press, New York